Temporal lobe epilepsy (TLE) is often associated with a pattern of neuropathology within the dentate gyrus (DG) that is strikingly similar to changes observed in several models of epilepsy. Granule cells (GCs, the principal cells of the DG) disperse, and their axons undergo extensive restructuring, sprouting into new terminal fields. Recent findings also indicate that after seizures, the neurogenesis of GCs increases, and some can migrate to the hilar/CA3 border (ectopic GCs). Determining how GC synaptic circuitry is altered is critical to understanding how hippocampal seizures could develop in TLE, since the DG is normally able to prevent electrographic seizures from spreading into the rest of the hippocampal formation. The proposed studies will test the hypothesis that following seizures, there is a selective strengthening of the pattern of synaptic connectivity among cells that could support recurrent excitation in the ventral DG (which is particularly excitable), thus promoting subsequent seizures. Tissue from control and experimental animals will be examined anatomically and physiologically four months after pilocarpine treatment, when spontaneous seizures have appeared. Aim I will ascertain if the synaptic input to, and output from, ectopic GCs is consistent with a role in a novel recurrent excitatory pathway. Dual electron microscopic (EM) immunolabeling techniques will be applied to characterize synaptic input to ectopic GCs in experimental tissue. Additionally, physiologically-identified and intracellularly-labeled ectopic GC axons in slices will be reconstructed at light and EM levels, after immunolabeling to identify neurons that receive output from ectopic GCs. Aim II will examine whether GC axons strengthen (in terms of synapse number and size) their innervation of hilar neurons that could support recurrent excitation (surviving mossy cells and ectopic GCs). This analysis will be conducted both across the whole population of terminals (by combining dual EM immunolabeling with stereological techniques), and within individual axons (by examining physiologically-identified and labeled axons from GCs in the GC layer). Since these fibers also contact interneuron subpopulations involved in recurrent inhibition, even small shifts in the balance of synaptic input could have a large impact on the excitability of the DG. The results of these studies will elucidate mechanisms underlying increased excitability in the DG, advancing our understanding of the pathophysiology of TLE.